IR-to-UV Optoelectronic Synapse for Object Identification
Abstract
Artificial Neural Networks (ANNs) have been implemented using software for object identification traditionally running on CMOS-based hardware.[1-4] NNs comprising artificial neurons and synapses which mimic their biological counterparts are efficiently realized using memristors.[5 10] Only recently, optoelectronic synapses are being realized to operate in visible/UV light [11] and at near-infrared (980 nm) wavelength [12]. In Year 2, we propose to implement the optoelectronic synapses that we developed in Year 1, in a neural network hardware to demonstrate its ability to identify images. We shall also work on improving the device’s detection range beyond 6 µm to enable its operation in MWIR-LWIR.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Mar 10, 2022
- Source ID
- FA86512210003
Entities
People
- Tania Roy
Organizations
- Air Force Research Laboratory
- United States Department of Defense
- University of Central Florida Board of Trustees